java.nio.file.FileSystemNotFoundException when getting file from resources folder

Question

I am getting this error on the following code (note that this does not happen on my local machine, only on my build server):

Files.readAllBytes(Paths.get(get         


        

Answer 1

Generally, it is not correct to assume that every resource is a file. Instead, you should obtain the URL/InputStream for that resource and read the bytes from there. Guava can help:

URL url = getClass().getResource("/elasticsearch/segmentsIndex.json");
String content = Resources.toString(url, charset);

Another possible solution, with the InputStream and apache commons: Convert InputStream to byte array in Java .

From a byte[], simply use the String constructor to obtain the content as a string.

Answer 2

If you use Spring, inject resources. Be it a file, or folder, or even multiple files, there are chances, you can do it via injection. Warning: DO NOT use File and Files.walk with the injected resources, otherwise you'll get FileSystemNotFoundException when running as JAR.

This example demonstrates the injection of multiple images located in static/img folder.

import org.springframework.beans.factory.annotation.Value;
import org.springframework.core.io.Resource;
import org.springframework.stereotype.Service;
import javax.annotation.PostConstruct;

@Service
public class StackoverflowService {

    @Value("classpath:static/img/*")
    private Resource[] resources;
    private List<String> filenames;

    @PostConstruct
    void init() {
        final Predicate<String> isJPG = path -> path.endsWith(".jpg");
        final Predicate<String> isPNG = path -> path.endsWith(".png");

        // iterate resources, filter by type and get filenames
        filenames = Arrays.stream(resources)
                .map(Resource::getFilename)
                .filter(isJPG.or(isPNG))
                .collect(Collectors.toList());
    }
}

Answer 3

You should get the resource through an InputStream and not a File, but there is no need for external libraries.

All you need is a couple of lines of code:

InputStream is = getClass().getResourceAsStream("/elasticsearch/segmentsIndex.json");
java.util.Scanner scanner = new java.util.Scanner(is).useDelimiter("\\A");
String json = scanner.hasNext() ? scanner.next() : "";

You can learn more about that method at https://stackoverflow.com/a/5445161/968244

Answer 4

You should be using getResourceAsStream(…) instead of getResource(…). There are a number of methods to read all bytes into a byte array, e.g. Apache Commons has a utility method to do just that.

Answer 5

Don't try to access a resource like a file. Just grab the InputStream and read the data from there:

byte[] data;
try (InputStream in = getClass().getResourceAsStream("/elasticsearch/segmentsIndex.json")) {
    data = in.readAllBytes​(); // usable in Java 9+
    // data = IOUtils.toByteArray(in); // uses Apache commons IO library
}

This example uses the IOUtils class from Apache commons-io library.

If you are targeting Java 9+ you can alternatively use data = in.readAllBytes​();.